Canadian Nosocomial Infection Surveillance Program

Final Report

Clostridium difficile Associated Diarrhea in Acute-Care Hospitals Participating in CNISP: November 1, 2004 to April 30, 2005

September 5, 2007

Prepared by: Denise Gravel Senior Epidemiologist Nosocomial and Occupational Infections Section Blood Safety and Surveillance, Health-Care Acquired Infections Division Public Health Agency of Canada Dr. Mark Miller Chair, Infection Prevention and Control SMBD–Jewish General Hospital Montreal, Quebec

2

Members of the Clostridium difficile Surveillance Working Group David Boyd National Microbiology Laboratory 1015 Arlington St. Winnipeg, MB R3E 3R2 Tel: 204-789-2133 Email: david_boyd@phac-aspc.gc.ca

Dr. Michael Gardam Toronto General Hospital 200 Elizabeth Street New Clinical Services Building, 12C-1261 Toronto, Ontario, Canada M5G 2C4 Tel: (416) 340 3758 Email: michael.gardam@uhn.on.ca

Denise Gravel (Co-chair) Nosocomial and Occupational Infections Public Health Agency of Canada 100 Eglantine Driveway, PL 0603E1 Ottawa, ON, K1A 0L9 Tel: (613) 841-3513 Email: denise_gravel@phac-aspc.gc.ca

Dr. Jim Hutchinson Health Care Corp. of St. John’s 300 Prince Philip Drive St. John’s, NL, A1B 3V6 Tel: (709) 777-7654 Email: jim.hutchinson@hccsj.nl.ca

Sharon Kelly Health Care Corp. of St. John’s 300 Prince Philip Drive St. John’s, NL, A1B 3V6 Tel: 709-777-3387 Email: sharon.kelly@hccsj.nl.ca

Dr. Allison McGeer Mount Sinai Hospital 1460-600 University Avenue Toronto, Ontario, M5G 1X5 Tel: (416) 586-4800 Ext 3118 Email: amcgeer@mtsinai.on.ca

Dr. Mark Miller (Co-chair) SMBD–Jewish General Hospital 3755 Cote St. Catherine, Suite G-139 Montreal, QC, H3T 1E2 Tel: (514) 340-8294 Email: mmiller@lab.jgh.mcgill.ca

Dr. Dorothy Moore Montreal Children’s Hospital 2300 Tupper, Room C1243 Montreal, QC, H3H 1P3 Tel: (514) 412-4485 Email: dorothy.moore@muhc.mcgill.ca

Dr. Michael Mulvey National Microbiology Laboratory 1015 Arlington St. Winnipeg, MB R3E 3R2 Tel: 204-789-2133 Email: michael_mulvey@phac-aspc.gc.ca

Dr. Andrew Simor Sunnybrook Health Sciences Centre Room B103, 2075 Bayview Avenue Toronto, ON, M4N 3M5 Tel: (416) 480-4549 Email: andrew.simor@sunnybrook.ca

Dr. Kathryn Suh Children’s Hospital of Eastern Ontario 401 Smyth Road Ottawa, ON, K1H 8L1 Tel: (613) 737-7600 ext. 2491 Email: ksuh@cheo.on.ca

Dr. Geoffrey Taylor University of Alberta Hospital 2E4.11 Walter McKenzie Centre 8440-112 Street Edmonton, AB, T6G 2B7 Tel: (780) 407-7786 Email: geoff.taylor@ualberta.ca

3

INTRODUCTION

Clostridium difficile associated diarrhea (CDAD) is the most frequent cause of nosocomial

infectious diarrhea in industrialized countries 1-3, affecting more than 300,000 hospitalized

patients yearly in the United States.4-5 Clinical manifestations range from asymptomatic

colonization, to severe diarrhea, pseudomembranous colitis, toxic megacolon and death.6

One of the earliest reports of more severe disease in patients with CDAD, many resulting

in death, was in Pittsburgh, Pennsylvania in 2000.7 Since the last half of 2002, several hospitals

in Quebec have experienced a dramatic increase in the incidence, severity and number of

relapses associated with CDAD.8-12 Similar reports have been seen in other industrialized

countries. 13, 14 An analysis of US hospital discharge data revealed that CDAD rates increased

abruptly beginning in 2001, with a doubling of national rates from 2000 to 2003.15 This increase

was most prominent for patients 65 years of age and older. Reports also suggested that the

attributable mortality rate (or fatality rate) had increased in recent years. Based on the data from

Quebec, the attributable mortality rate for CDAD was estimated at 6.9%. 9

Shortly after the appearance of reports of more severe disease in patients with CDAD, a

previously unknown strain of C.difficile was identified. 9, 16 The strain is characterized as North

American pulsed-field Type 1 with a restriction enzyme analysis type BI and PCR ribotype 027,

hence the name NAP1/B1/027 (more commonly referred to as NAP1/027 or simply NAP1). In

addition to the clostridial enterotoxin A and cytotoxin B, the main virulence factors of C.difficile,

NAP1/BI/027 possesses an extra toxin known as the binary toxin. The role of this toxin remains

unclear. More importantly, this strain has been shown, in vitro, to produce greater quantities of

toxins A and B; due to an 18-base pair tcdC gene deletion. 9, 13, 16, 17 Although it is not known if

these unique characteristics are responsible for the increased virulence, studies have found a

definite association between NAP1/BI/027 and more severe disease, especially in older patients

with CDAD. 18

In this context, the Canadian Nosocomial Infection Surveillance Program (CNISP) elected

to re-examine the incidence of CDAD in Canada with an emphasis on patient outcomes. The

4

objectives of the surveillance were to: 1) determine the incidence and burden of illness

associated with CDAD in CNISP hospitals; 2) determine if there is an increase in severe

outcomes (mortality and morbidity associated with CDAD) in 2005 compared to 1997; 3)

characterize molecular subtype/toxinotype of C.difficile strains and determine if certain strains are

associated with severe clinical outcomes; and 4) determine the geographic distribution of

C.difficile isolates, including NAP1/027.

BACKGROUND

1997 N-CDAD Point Prevalence Surveillance Project

In 1997, the CNISP conducted a six week prospective surveillance study for healthcare-

associated CDAD (HA-CDAD, formerly termed N-CDAD) within 19 health care hospitals in 8

Canadian provinces.19 During this period; the participating hospitals tested all diarrheal stools

from hospitalized patients for C.difficile toxin detection. Questionnaires were completed for

patients with positive C.difficile assays who met eligibility criteria (diarrhea >2 days, symptoms

occurred 3 days or more after admission, or symptoms causing readmission within one month of

the current admission).

Among inpatients with diarrheal stools, 13% were caused by C.difficile. The incidence of

healthcare-associated CDAD (HA-CDAD) cases was 6.63 cases per 10,000 patient days (95% CI

3.75-9.51) and 5.9 cases per 1000 patient admissions (95% CI 3.4-8.4). CDAD was found most

frequently in older patients and those hospitalized more than 2 weeks in medical or surgical

wards.

A sub-section of the initial project addressed morbidity, mortality and healthcare burden

of healthcare- acquired CDAD in the same centers. Of the 269 patients that satisfied the HA-

CDAD case definition, 41 (15.2%) died, 4 (1.5%) of these were attributable to CDAD. 20 The

annual cost of HA-CDAD readmission for each center was estimated to be at least of $128,200.

These reports were pivotal since they provided baseline rates to which other Canadian hospitals

could compare and provided the only available information on healthcare burden of CDAD on

5

Canadian hospitals.

Increase in incidence, severity and relapse in Canadian hospitals

Recent reports have suggested an increase in incidence, severity and/or risk of relapse

of CDAD in Canada. 8 Several health-care institutions in Quebec (located mostly in the southern

region of the St-Lawrence River, Montreal and the Eastern Townships) are reporting increased

incidence of nosocomial healthcare-associated cases, with average rates of 25 cases per 1000

admissions. 21

An increase in the frequency of C.difficile toxin-producing strains at the Centre Hospitalier

Universitaire de Sherbrooke lead to a 13 year review (January 1991-December 2003) of 1721

cases of CDAD in the Eastern Townships of Quebec. 12 From 1991/92 to 2001/02, the overall

annual incidence of CDAD was stable with rates of 35 per 100,000 to 50 per 100,000. However,

in 2003 the incidence increased to 160 per 100,000 with a rate of almost 900 per 100,000 for

adults older than 65.

Severe colitis was defined as perforation, toxic megacolon, shock or death within 30 days

of diagnosis. Using this definition, in 2003, an increase in the incidence of severe colitis when

compared to previous years 1991-2002 (adjusted OR 2.2; 95% CI, 1.0-4.9) was reported.

Further study, conducted in 2005 in 88 Quebec hospitals found that severe disease was

twice as frequent among patients with C.difficile strains possessing binary toxin genes and tcdC

deletion compared to patients with strains lacking these characteristics. 18 This study was the

first to describe the geographic dissemination of NAP1 strain in Quebec.

Starting in early 2006, a number of media reports have described outbreaks in areas

outside of the Montreal region and in Ontario associated with severe disease in patients with

CDAD; notably Gatineau, Quebec and Sault-Ste-Marie, Mississauga, and Belleville, Ontario. 21-23

To date, there have been no reported outbreaks in the western provinces or the Atlantic region.

Canadian Nosocomial Infection Surveillance Program

The CNISP is a collaborative effort of the Canadian Hospital Epidemiology Committee

6

(CHEC), a subcommittee of the Association of Medical Microbiologists and Infectious Disease

(AMMI) and the Centre for Infectious Diseases Prevention and Control (CIDPC) of the Public

Health Agency of Canada.

Established in 1994, the objectives of CNISP are to provide rates and trends on

healthcare-associated (formerly nosocomial) infections at Canadian health care facilities thus

enabling comparison of rates (bench-marks), and providing evidence-based data that can be

used in the development of national guidelines on clinical issues related to healthcare-associated

infections. At present, 49 sentinel hospitals from 9 provinces participate in the CNISP network.

All CNISP hospitals have a university affiliation and provide primary, secondary, and tertiary care

to adult and/or pediatric patients. Seven hospitals are stand-alone pediatric centers.

CHEC members participate in CNISP by working on sub-committees that direct the

development, implementation and analysis of surveillance projects. CHEC members and their

corresponding healthcare institution(s) participate voluntarily in CNISP projects by collecting

standardized, case-by-case, non-nominal data on hospitalized patients at risk of healthcare-

associated infections. The data is submitted to PHAC for compilation and analysis. All data is

analyzed by region or larger geographical area. At no time is submitted data analyzed by

individual hospital or site. The results of the CNISP surveillance projects are disseminated

primarily through publications in peer-reviewed medical journals. The PHAC employees

participate by co-authoring the manuscripts along with CHEC.

Members of the CNISP have collaborated successfully on a number of other surveillance

projects including surveillance for methicillin-resistant Staphylococcus aureus (MRSA),

Vancomycin- resistant Enterococcus (VRE), and hemodialysis-associated blood stream

infections, to name only a few. Although the patient populations examined in our surveillance

activities are in major teaching hospitals and so likely not entirely representative of all hospitalized

patients in Canada, the data obtained from our surveillance provide an important contribution to

understanding the impact of healthcare-associated infections in patients admitted to Canadian

hospitals.

7

METHODS

A prospective surveillance for CDAD was conducted among the patients hospitalized in

Canadian acute-care hospitals participating in CNISP between November 1, 2004 and April 30,

2005. A total of 34 hospitals participated in the surveillance activity including one non-CHEC

hospital from the province of Ontario. Of these, 16 hospitals admitted pediatric patients under

the age of 18 years. All hospitalized patients 1 year of age and older meeting the case definition

for CDAD were eligible for enrolment. Both community-acquired (CA-CDAD) and healthcare-

associated (HA-CDAD) cases were included.

The following case definition was utilized for CDAD: 1) either diarrhea over 2 days; or

fever, abdominal pain and/or ileuses with laboratory confirmation of a positive toxin assay for

C.difficile; or 2) diagnosis of pseudomembranous colitis on colonoscopy, or

histological/pathological diagnosis of CDAD. The infection was considered healthcare-associated

if the patient’s symptoms occurred at least 72 hours after admission; or symptoms resulted in

readmission of a patient who had been hospitalized within the previous two months of the current

admission date, and who was not a resident in a chronic care facility or nursing home. Patients

who met the case definition for CDAD but did not meet the surveillance definition for healthcare-

associated CDAD were considered to have acquired CDAD in the community.

Eligible patients were identified by daily review of stool C.difficile toxin assay results in

the clinical microbiology laboratory as well as a review of relevant pathology reports or operating

room records. The charts of patients with positive stool samples for C.difficile toxin were

examined by experienced and trained infection control professionals or trained research

personnel associated with each hospital. Basic demographic data was collected on all patients,

including age and sex, admission date, type of patient ward where the patient was located on the

day CDAD was identified, medical service, primary admitting diagnosis and co-morbidities.

Information about CDAD included date of onset of diarrhea and date of the first positive specimen

submission, initial treatment of CDAD, medical and treatment inventions, and complications.

8

Data regarding adverse events was collected 30 days after the diagnosis of a positive

case and included death (all cause, and attributable to CDAD), ICU admission, surgery, bowel

perforation, GI bleeding, toxin megacolon, dehydration, hypokalemia, and relapse. A relapse was

defined as an episode of illness with onset of symptoms within 2 months of the end of the

previous CDAD episode. All cases of death within 30 days of diagnosis of a CDAD episode were

assessed by the CHEC member or a designated physician to determine if the death was

attributable to CDAD. Cause of death was determined by the following criteria: 1) CDAD was

directly related to the death of the patient; that is, the patient had no other underlying condition

that would have caused death during this hospitalization; or 2) CDAD was indirectly related to

death; that is, the CDAD contributed to the patient’s death but was not the primary cause; or 3)

the patient died but CDAD was not related to death.

Data was recorded for each new CDAD episode. An episode was defined as the time

from onset of symptoms until the last day of antibiotic treatment. Episodes which did not meet

the criteria for disease (i.e., positive test, but criteria for diarrhea or other symptoms not met), and

relapses, were not included in the surveillance. Episodes occurring more than 2 months after the

first episode were considered new episodes.

Data were collected and entered manually onto patient data extraction forms and

forwarded to the PHAC for data entry and analysis. Information from the patient questionnaires

was entered at the PHAC on a web-based data entry system (WEBBS) that mirrored the patient

questionnaire. A unique identifier linked to the patient name was used only to identify patients at

the participating hospital and was not transmitted to the PHAC. While this surveillance project

was observational and did not involve any alteration in patient care, ethics approval was obtained

at some of the participating hospitals.

Whenever possible, frozen stool specimens from patients with CDAD were forwarded to

the National Microbiology Laboratory (NML) in Winnipeg for molecular characterization. C.difficile

was isolated from the stools using an alcohol shock procedure. Toxigenic strains were confirmed

using polymerase chain reaction (PCR) to detect tcdA and tcdB genes. PCR was also used to

9

confirm the species, detect variations in the tcdC gene and to detect the presence of the binary

toxin (cdtB). Pulsed-field gel electrophoresis (PFGE) was used to type the strains. Antimicrobial

susceptibilities to 12 antimicrobials were determined using agar dilution following Clinical

Laboratory Standards Institute (CLSI) guidelines. Resistant breakpoints used were as follows:

metronidazole >16 mg/L, vancomycin >16 mg/L, teicoplanin >16 mg/L, clindamycin >4 mg/L,

ciprofloxacin >4 mg/L, levofloxacin >4 mg/L, gatifloxacin >4 mg/L, moxifloxacin >4 mg/L, cefazolin

>16 mg/L, cefuroxime >32 mg/L, ceftriaxone >32 mg/L and cefotaxime >32 mg/L.

Statistical analysis

Incidence and rates of CDAD were calculated by province or region using patient

admissions and patient-days for denominator. Mortality and case fatality rates were determined

using the criteria described in the ‘methods’ section. Descriptive and univariate analyses were

performed. To assess differences between patient populations, continuous variables were

expressed by means and compared using the Student t-test and/or the Mann-Whitney test.

Hospital stay was expressed by median and interquartile rank. Categorical variables were

expressed as proportions and compared using the chi-square test and the Fisher’s t-test when

necessary. All tests were two-tailed, and a P value of less than 0.05 was considered statistically

significant. Relative risks with corresponding 95% confidence intervals were calculated according

to standard methods. Multivariate logistic regression model will be used to assess patient factors

associated with a severe outcome. Severe outcome was defined as an admission to the

intensive care unit for complications of CDAD, colectomy, and/or death, directly or indirectly

related to CDAD. Variables will be selected for entry into the regression model if at least 10 of

the patients had the characteristic and the variables were significantly associated with a severe

outcome at a p-value less than or equal to 0.25 in the univariate analysis. The goodness-of-fit of

the final model will be tested using the deviance test. Statistical analysis was conducted using

SAS version 9.1 (SAS Institute, Cary NC).

RESULTS

10

Incidence and rates of CDAD

A total of 1842 patients with primary or recurrent CDAD were identified during the 6-

month surveillance period. Of these, 1745 (95%) were adults 18 years of age and older and 97

(5%) were children 1 to 17 years of age. The source of the CDAD infection was found to be

healthcare-associated (HA-CDAD) in 1493 (81%) of the patients and community-acquired (CA-

CDAD) in 292 (16%) of the patients. Fifty-seven (3%) patients acquired the infection in a nursing

home (Table 1.)

The overall national incidence and rate of HA-CDAD for during this 6-month period was

4.5 cases per 1000 patient admissions and 6.4 per 10,000 patient-days (Table 2). The incidence

and rate were significantly higher in Quebec than the rest of Canada (11.1 vs. 3.9 cases per 1000

admissions and 11.9 vs. 5.7 cases per 10,000 patient-days, p < 0.0001). The rates are similar to

those found in our previous study; 6.4 vs. 6.6 cases per 10,000 patient-days in 1997.

At 30 days following onset of CDAD, 237 patients had died from all causes for a mortality

rate of 15.9 per 100 cases. Of these, 84 were directly (31 deaths (2.1%)) or indirectly (53 deaths

(3.6%)) related to CDAD for a case fatality rate of 5.6% (Table 3). The fatality rate in Quebec

was four times higher than the rest of the Canada combined (14.8% vs. 3.5%, p < 0.0001).

Compared to the surveillance performed in 1997, the deaths directly or indirectly related to CDAD

increased by almost 400% (5.6% vs. 1.5% in 1997, p < 0.0001). There were 4 deaths in children

1 to 17 years of age; 2 deaths were indirectly related to CDAD.

The data presented below describes the results found in adult patients 18 years of age

and older with HA-CDAD.

HA-CDAD in Canadian adults

Of the total 1842 patients with CDAD, 1493 (81%) patients had HA-CDAD. Of these,

1430 (96%) were adults 18 years of age and older and 63 (4%) were children between 1 and 17

years of age (Table 1). The mean age of the adults was 70 ±16 years (range 18-101); 996 (70%)

were 65 years of age and older; 735 (51%) were males (Table 4). The majority of the patients,

1242 (87%) were acute-care patients while the remaining 188 (13%) were long-term care

11

patients. The mean length of stay before onset of CDAD was 25 ± 50 days (median: 11 days).

There was no difference in the mean length of stay between the adults 65 years of age and older

and the adults 18 to 64 years of age.

At the time of the onset of HA-CDAD, 609 (43%) patients were on medical wards, 327

(23%) patients were on a surgical unit, and 142 (10%) patients were in an Intensive Care Unit

(ICU). The remaining 352 (24%) were on other wards including oncology/hematology, long-term

care wards and transplant units.

Patients 65 years of age and older were more likely to acquire CDAD on medical ward (48% vs.

30%, p < 0.0001), whereas the adults 18 to 64 years of age were more likely to have acquired

CDAD on a surgical or oncology/hematology unit (28% vs. 21%, p = 0.001 and 7% vs. 2%, p <

0.0001; respectively).

Only 81 (6%) patients did not receive treatment for the episode of CDAD. Among the

1430 patients with CDAD, 1215 (85%) were prescribed Metronidazole, 230 (16%) received

Vancomycin and 51 (4%) received Probiotics (Table 5). A total of 168 (12%) patients were

receiving more than one drug. Patients 65 years of age and older were 1.5 times more likely to

receive Vancomycin than patients aged 18 to 64 years (18% vs. 12%, p = 0.0053). Of interest,

patients with CDAD in the province of Quebec were 9 times more likely to receive Vancomycin

than all other provinces and regions combined (56% vs. 6%, p < 0.0001) (Table 6).

Adverse outcomes in patients with HA-CDAD

A total of 319 (22%) adult patients with HA-CDAD developed complications in the first 30

days following onset of CDAD; 104 (7.3%) patients had a severe outcome. Relapse was the

most common complication occurring in 125 (9%) patients. Dehydration occurred in 84 (6%)

patients; hyopkalemia was seen in 39 (3%) patients; and pseudomembranous colitis and/or

gastrointestinal bleed not requiring transfusions was seen in 47 (3%) patients. Dehydration was

more likely to be seen in the elderly patients 65 years and older (7% vs. 3%, p = 0.005) whereas

gastrointestinal bleeding requiring blood transfusions were more likely to be seen in adults aged

18 to 64 years (2% vs. 0.7%, p = 0.029).

12

Thirty-one (2%) patients were admitted to the ICU for complications related to CDAD and

12 (1%) patients underwent a colectomy. In addition, 82 patients died, either directly or indirectly

related to CDAD for a case fatality rate of 5.7%. The case fatality rate was 3.5 times higher in

patients over the age of 65 years compared to the patients aged 18 to 64 years (7.3% vs. 2.1%, p

< 0.0001).

In univariate analysis the following factors were associated with a severe outcome: age

65 years and older (RR 2.23, 95% CI 1.34-3.70, p = 0.001); being a long term care patient on any

ward, having been admitted from a long term care facility or having acquired CDAD on a long

term care unit (RR 1.77, 95% CI 1.14-2.78, p =0.01; RR 2.15, 95% CI 1.34-3.45, p = 0.002; and

RR 1.97, 95% CI 1.01-3.86, p = 0.05, respectively); dementia (RR 1.88, 95% CI 1.10-3.24, p =

0.02); having received Vancomycin as initial treatment for CDAD (RR 2.60, 95% CI 1.97-3.45, p <

0.0001); and having had a change in the initial treatment for CDAD (RR 2.77, 95% CI 1.90-4.03,

p < 0.0001) (Table 8). When compared to patients on all other units, patients on surgical units

were less likely to have a severe outcome (RR 0.44, 95% CI 0.24- 0.79, p < 0.004). In addition,

patients receiving Metronidazole as initial treatment were also less likely to have a severe

outcome (RR 0.84, 95% CI 0.74-0.95, p < 0.0001).

Laboratory characterization of C.difficile isolates

During the 6-month surveillance, 2307 frozen stool specimens were submitted to NML for

identification and characterization. Of these, 450 were duplicate specimens and were discarded.

Laboratory analysis was completed on 1857 stool specimens; 488 of the 1857 were from patients

who did not meet the case definition for CDAD and for which no patient information was collected.

The laboratory data from the remaining 1369 was linked to the clinical database.

PFGE patterns of C.difficile isolates: HA-CDAD in adults

Of the 1430 patients with HA-CDAD, toxigenic C.difficile was found in the stool

specimens of 1008 (70%) patients. In 280 (20%) patients, a stool specimen was not submitted

to NML. In 101 (7%) patients, C.difficile was not isolated and in 41 (3%) non-toxigenic C.difficile

was found. These patients all had a positive toxin assay as determined by the hospital

13

laboratory. The overall recovery rate for C.difficile was 91%.

Among the 1008 patients with complete laboratory workup, the hypervirulent strain

NAP1/027 was isolated in 311 (31%) patients (Table 9). NAP1 was 1.3 times more likely to be

found in patients 65 years of age and older compared to adults, 18 to 64 years of age (33% vs.

25%, p = 0.006) whereas NAP4 was more likely to be seen in younger adults (6% vs. 3%, p =

0.003). The second most predominant NAP subtype was NAP2, commonly termed “J-strain”.

The strain was recovered in the stool specimens of 283 (28%) patients with HA-CDAD. All other

strains, not classified as a NAP subtype, accounted for the majority of the C.difficile strains

isolated; over 115 distinct PFGE patterns were found. In addition, 316 (31%) of the strains were

binary toxin positive and tcdC gene deletion positive; the vast majority of these were NAP1 (296

(29%)).

The NAP1/027 strain was found in every province and region of Canada (Table 10).

The NAP1/027 was the most prominent strain in Quebec, isolated in 171 (76%) of the adult

patients with HA-CDAD. The NAP1/027 strain was seen in 100 of 451 (22%) patients in Ontario

and 22 of 111 (20%) patients in Alberta (Figure 1). The Atlantic region had the lowest incidence

of NAP1/027, with the isolate found in only 5 of 103 (5%) of the adult patients with HA-CDAD.

The most prominent NAP subtype in the Atlantic region was NAP2; isolated in 64 of 103 (62%)

patients. This subtype was also the most frequently isolated NAP subtype in Ontario; found in

157 of 451 (35%) patients.

Antibiotic susceptibility of C.difficile: HA-CDAD in adults

Among the 1008 strains isolated in adults with HA-CDAD, there were no strains resistant

to Metronidazole, Vancomycin, and Teicoplanin. All strains were resistant to Ciprofloxacin,

Cefuroxime and Cefotaxime (Table 11). A total of 868 (86%) strains were resistant to

Clindamycin, 957 (95%) resistant to Cefazolin and 747 (74%) resistant to Levofloxacin. The

NAP1 strain was found more likely to be resistant to the fluoroquinolones than the other strains

combined (Levofloxacin, 92% vs. 66%, p < 0.001; Gatifloxacin, 83% vs. 60%, p < 0.001 and

Moxifloxacin, 83% vs. 60%, p < 0.001) whereas the NAP1 was more likely to sensitive to

14

Clindamycin (82% vs. 88%, p = 0.02).

Clinical outcome and presence of NAP1/027 strain

Adult patients with HA-CDAD were found to be twice as likely to develop a severe

outcome if the NAP1/027 strain was isolated in their stool or die directly or directly related to the

CDAD (12.5% vs. 5.9%, RR 2.12, 95% CI 1.65-2.59, p = 0.0003 and 12.5% vs. 5.9%, RR 2.34,

95% CI 1.50-4.11, p < 0.0001, respectively). The effect of the strain type was significantly

associated with the age of the patient (Figure 2). In patients under the age of 60 years, strain

type did not seem to be associated with severe outcomes. Over the age of 60 years (but not in

the extreme elderly > 90 years of age), infection with NAP1/027 was highly associated with

severe outcomes (p = 0.03). In the extreme elderly ( ≥ 90 years of age), severe outcomes were

frequent, regardless of the strain type.

SUMMARY

The results from this surveillance project represent the most comprehensive surveillance

of CDAD in Canada. This demonstrates that coordinated national attempts to survey CDAD, and

more specifically, HA-CDAD; can be accomplished with excellent patient information and isolate

recovery, given sufficient planning time and resources, to create a linked clinical-microbiological

CDAD database. The information contained in this national database also represents the largest

such database to exist in the developed countries including the US.

Our study demonstrates the wide variations in HA-CDAD among the participating

hospitals. The underlying reasons remain as yet unclear although previous studies have

suggested that antibiotic usage; the physical layout of the institution including the presence or

absence of sinks for hand washing; and the infection prevention and control practices including

isolation practices have played a role in the overall incidence of HA-CDAD. 3, 25, 26 Wide

variations are also seen between provinces and regions in Canada. Ontario was found to have

40% more HA-CDAD cases (5.7 vs. 3.8 cases per 1000 admissions and 7.8 vs. 5.5 per 10,000

patient days, p < 0.0001); and Quebec has over twice the incidence of HA-CDAD compared to

15

other areas of the country (11.1 vs. 3.9 cases per 1000 admissions and 11.9 vs. 5.7 per 10,000

patient days, p < 0.0001). Overall, there is a small decrease in the mean incidence of HA-CDAD

in Canada since 1997; however our surveillance has found a significant increase in the number of

deaths related to CDAD and severe outcomes. Compared to the CNISP surveillance conducted

in 1997, the incidence of deaths directly or indirectly related to CDAD has increased almost 4-fold

(5.6 vs. 1.5%, p < 0.0001). 20 These results are comparable since we used the same

methodology with this surveillance as was utilized in 1997. Furthermore, we found that the case

fatality rates from CDAD were much higher in Quebec, followed by Ontario. Our findings support

previously published reports describing increased fatality in Quebec. 18

The presence of the NAP1/027 strain closely mirrors the HA-CDAD incidence and severe

outcomes, across all provinces and regions. The “hypervirulent” NAP1/027 strain was found in

eight provinces, but mostly in British Columbia, Alberta, Ontario, and Quebec. In our study, we

found that the NAP1/027 strain leads to severe outcomes more frequently (approximately 3 times

the incidence) in adults 60 to 90 years of age. Adults aged 90 and over were found to have a 14

to 16% attributable mortality, regardless of the strain. This data is remarkably similar to the

Quebec outbreak data, in terms of the incidence of age-related CDAD attributable mortality. 18

There are limitations to our study, primarily inherent to large multi-centre surveillance

activities. First, although data collection was conducted by experienced and trained infection

control professionals using standardized definitions, the data collection remained unmonitored

and there may be inconsistencies between hospitals in identifying a CDAD. As the diagnosis of

a CDAD is frequently based on laboratory findings, there may be some variability in the

microbiological laboratory testing and identification of C.difficile at the different hospitals.

Seasonal variations in CDAD incidence could influence the overall annual rates. Our surveillance

was conducted during the peak CDAD months of November to April. In addition, although the

hospital epidemiologist or another qualified physician determined the cause of death in patients

with HA-CDAD; attribution of mortality; i.e. direct or indirect, is always subjective and can be

interpreted differently from one clinician to another. Finally, the populations examined in this

16

survey were in major teaching hospitals and so likely not entirely representative of all hospitalized

adult patients in Canada.

Despite these limitations, the data presented in this study are an important

contribution to understanding the impact of CDAD in adults admitted to Canadian hospitals. The

results are sufficiently robust to be used as baseline indicators for future comparisons. Follow-up

surveillance in the same hospitals will allow us to follow the spread of C.difficile strains; more

specifically the spread of NAP1/027, in Canada and assess the impact on the morbidity and

mortality associated with CDAD. National surveillance also provide opportunities for interhospital

collaboration that may lead to more standardized use of surveillance methodology, including

application of definitions and case finding methods, and effective infection prevention and control

measures.

NEXT STEPS

A multivariate logistic regression will be undertaken to evaluate the association between

the variables and severe outcome and/or death related to HA-CDAD. Additional sub-analysis is

currently underway, including (but not limited to): description of the pediatric patients between 1

and 18 years of age with CDAD; description of the CA-CDAD cases and case of CDAD in

patients residing in nursing homes; and an evaluation of the C.difficile isolates from patients who

did not meet the case definition for CDAD to determine if the strains are different than the strains

isolated from patients who met the case definition for CDAD. The results in this report will be

presented at the 47th Annual Interscience Conference on Antimicrobial Agents and

Chemotherapy (ICAAC) Conference in Chicago, September 17- 20, 2007 and the 45th Infectious

Disease Society of America (IDSA) Annual Meeting in San Diego, October 4-7, 2007.

Ongoing surveillance for CDAD

As of January 1st, 2007, surveillance for CDAD will be ongoing and mandatory in all

hospitals participating in CNISP. CDAD surveillance will consist of monthly reporting of

incidence and rates, and an annual two-month targeted surveillance for patient outcomes and

17

laboratory characterization of C.difficile isolates. The numerator will consist of all hospitalized

patients meeting the case definition for CDAD except for children less than 1 year of age and

patients residing on psychiatric and long-term care or awaiting placement units. Both community-

acquired and hospital-acquired cases of CDAD will be counted. Denominator information will

include total patient days, patient admissions, and total number of liquid stools submitted to

microbiology laboratory. Whenever possible, denominator information will be obtained

separately for pediatric and adult patients. The information received will be tabulated every four

months and disseminated through the Nosocomial and Occupational Infections website at

www.nosocomial.ca.

Once a year over a two-month consecutive period, a patient data collection form will be

completed for on all patients meeting the case definition for CDAD and frozen stool specimens

will be forwarded to NML for characterization. The annual period for the targeted surveillance

will be determined each year at the CNISP meeting for the next calendar year. The targeted

surveillance for 2007 will be from March 1st to April 30th. Patient information will include: age

and sex, admission date, and type of patient ward where the patient is located on the day CDAD

is identified. Information about CDAD will include date of onset of diarrhea or date of the first

positive specimen submission. Data regarding adverse events will be collected 30 days after the

diagnosis of a positive case, and will include death (all cause, and related to CDAD), ICU

admission, and colectomy.

(2007 CNISP surveillance protocol available upon request)

ACKNOWLEDGEMENTS

The CDAD working group acknowledges the contribution of the following individuals who assisted

with project management, data collection and data entry: Melinda Piecki, Katie Cassidy, John

Koch, Emma Ongsansoy, Monali Varia, Mark Osmond, Stephanie Leduc, Louis Valiquette, and

the Infection Control Professionals in each participating hospital.

18

REFERENCES

1. McFarland LV. Epidemiology of infectious and iatrogenic nosocomial diarrhea in a cohort of general medicine patients. Am J Infect Control 1995; 23(5): 295-305.

2. Kelly CP, Pothoulakis C, LaMont JT. Clostridium difficile colitis. N Eng J Med 1994;

330(4): 257-62.

3. Poutanen SM, Simor AE. Clostridium difficile-associated diarrhea in adults. CMAJ 2004; 171(1): 51-8.

4. McFarland LV, Mulligan ME, Kwok RY, Stamm WE. Nosocomial acquisition of

Clostridium difficile infection. N Engl J Med 1989; 320(4): 204-10.

5. Johnson S, Clabots CR, Linn FV, Olson MM, Peterson LR, Gerding DN. Nosocomial Clostridium difficile colonization and disease. Lancet 1990; 336(8707): 97-100.

6. Kelly CP, LaMont JT. Clostridium difficile infection. Annu Rev Med 1998; 49: 375-90.

7. Muto CA, Pokrywka M, Shutt K, et al. A large outbreak of Clostridium difficile –

associated disease with an unexpected proportion of deaths and colectomies at a teaching hospital following increased Fluoroquinolone use. Infect Control Hosp Epidemiol 2005; 26: 273-80.

8. Valiquette L, Low PE, Pepin J, McGeer A. Clostridium difficile infection in hospitals: a

brewing storm. CMAJ 2004; 171(1): 27-9.

9. Loo VG, Poirier L, Miller MA, et al. A predominatly clonal multi-institutional outbreak of Clostridium difficile –associated diarrhea with high morbidity and mortality. N Engl J Med 2005; 2442-9.

10. Eggertson L, Sibbald B. Hospitals battling outbreaks of Clostridium difficile. CMAJ 2004;

171(1): 19-21.

11. Louie TJ, Meddings J. Clostridium difficile infection in hospitals: risk factors and responses. CMAJ 2004; 171(1): 45-6.

12. Pepin J, Valiquette L, Alary ME, et al. Clostridium difficile – associated diarrhea in a

region of Quebec from 1991-2003: a changing pattern of disease severity. CMAJ 2004; 171(5): 466-72.

13. Warny M, Pepin J, Fang A, et al. Toxin production by an emerging strain of Clostridium

difficile associated with outbreaks of severe disease in North America and Europe. Lancet 2005; 366: 1079-84.

14. Kuijper EJ, Coignard B, Tull P. Emergence of Clostridium difficile – associated disease in

North America and Europe. Clin Microbiol Infect 2006; 12 (Suppl 6): 2-18.

15. McDonald LC, Owings M, Jernigan D. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996-2003. Emerg Infect Dis 2006; 12: 409-15.

16. McDonald LC, Killgore GE, Thompson A, et al. An epidemic, toxin gene-variant strain of

Clostridium difficile. N Engl J Med 2005; 353: 2433-41.

19

17. Blossom DB, McDonald LC. The challenges posed by reemerging Clostridium difficile infection. CID 2007; 45(25): 222-227.

18. Hubert B, Loo VG, Bourgault AM et al. Portrait of the geographic dissemination of the

Clostridium difficile North American pulsed-field type 1 strain and the epidemiology of C.difficile – associated disease in Quebec. Clin Infect Dis 2007; 44: 238-244.

19. Hyland M, Ofner-Agostini M, Miller M, Paton S, Gourdeau M, Ishak M, N-CDAD in

Canada: Results of the Canadian Nosocomial Infection Surveillance N-CDAD Prevalence Surveillance Project. Can J Infect Dis 2001; 12(2): 81-8.

20. Miller MA, Hyland M, Ofner-Agostini M, Gourdeau M, Ishak M. Morbidity, mortality, and

healthcare burden of nosocomial Clostridium difficile – associated diarrhea in Canadian hospitals. Infect Control Hosp Epidemiol 2002; 23(3): 137-40.

21. Loo VG, Libman MD, Miller MA, et al. Clostridium difficile: a formidable foe. CMAJ 2004;

171(1): 47-8. 22. The Toronto Star: Lethal germ hits hospital; Bacteria strain behind 2,000 deaths in

Quebec shows up in Mississauga, where 4 patients have died, 14 are treated. March 1, 2007.

23. The Globe and Mail. C.difficile widens its reach, Quebec official says. November 11,

2006.

24. The Globe and Mail. How bad is the C.difficile problem in Canadian hospitals? Is it getting worse? Better? No one really knows. Sure, there are anecdotal reports of outbreaks in individual hospitals, many of which keep close tabs on their in-house C.difficile. May 2, 2007.

25. McFarland LV, Beneda HW, Clarridge JE, Raugi GJ. Implications for the changing face of

Clostridium difficile disease for health care practitioners. Am J Infect Control 2007; 35: 237-53.

26. Blossom DB, McDonald LC. The challenges posed by reemerging Clostridium difficile

infection. CID 2007; 45: 222-7.

20

CNISP Hospitals that participated in the surveillance for CDAD; November 1, 2004 to April 30, 2005*

Victoria General Hospital Victoria, British Columbia Vancouver General Hospital Vancouver , British Columbia

Peter Lougheed Centre (3 facilities) Calgary , Alberta

University of Alberta Hospital Edmonton , Alberta

Stollery Children’s Hospital Edmonton, Alberta

Health Sciences Centre (2 facilities) St-John's, Newfoundland

QE II Health Sciences Centre Halifax , Nova Scotia

I.W.K. Hospital for Sick Children Halifax , Nova Scotia

Hospital for Sick Children Toronto , Ontario

Mount Sinai Hospital Toronto , Ontario

St-Joseph’s Health Centre Hamilton , Ontario

Peterborough General Hospital Peterborough, Ontario

Children’s Hospital of Eastern Ontario Ottawa, Ontario

Health Sciences Centre (Adults) Winnipeg , Manitoba

Hamilton Health Sciences (4 facilities) Hamilton , Ontario

Health Sciences Centre (Paediatrics) Winnipeg , Manitoba

London Health Sciences Centre (2 facilities) London , Ontario

The Ottawa Hospital Ottawa , Ontario

Royal University Hospital (2 facilities) Saskatoon , Saskatchewan

Jewish General Hospital Montréal, Québec

Montreal Children's Hospital Montreal , Quebec

Maisonneuve-Rosemont Hospital Montreal, Quebec

Sunnybrook and Women's College Health Science Centre Toronto , Ontario

The Moncton Hospital Moncton , New Brunswick

Kingston General Hospital Kingston , Ontario

* 34 hospitals participated in the surveillance. Some CHEC sites have more than one facility.

21

Members of the Canadian Nosocomial Infection Surveillance Program as of January 1st, 2007. David Boyd, National Microbiology Laboratory, Public Health Agency of Canada; Elizabeth Bryce Vancouver General Hospital, Vancouver, BC; John Conly, Foothills Medical Centre Calgary, Alta; Gordon Dow, The Moncton Hospital, Moncton, NB; John Embil, Health Sciences Centre Winnipeg, Man; Joanne Embree, Health Sciences Centre, Winnipeg, Man; Sarah Forgie, Stollery Children’s Hospital, Edmonton, Alta; Charles Frenette, Hôpital Charles LeMoyne, Longueil, Que; Michael Gardam, University Health Network, Toronto, Ont; Denise Gravel, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada; Elizabeth Henderson, Peter Lougheed Centre, Calgary, Alta; James Hutchinson, Health Sciences Centre, St. John’s, Nfld; Michael John, London Health Sciences Centre, London, Ont; Lynn Johnston, Queen Elizabeth II Health Sciences Centre, Halifax, NS; Pamela Kibsey, Victoria General Hospital, Victoria, BC; Joanne Langley, I.W.K.Health Science Centre, Halifax, NS; Mark Loeb, Hamilton Health Sciences Corporation, Hamilton, Ont; Anne Matlow, Hospital for Sick Children, Toronto, Ont; Allison McGeer, Mount Sinai Hospital, Toronto, Ont.; Sophie Michaud, CHUS-Hôpital Fleurimont, Sherbrooke, Que; Mark Miller, SMBD-Jewish General Hospital, Montreal, Que; Dorothy Moore, Montreal Children’s Hospital, Montreal, Que; Michael Mulvey, National Microbiology Laboratory, Public Health Agency of Canada; Marianna Ofner-Agostini, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada; Shirley Paton, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada; Virginia Roth, The Ottawa Hospital, Ottawa, Ont; Andrew Simor, Sunnybrook and Women’s College Health Sciences Centre, Toronto, Ont; Jacob Stegenga, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada; Tammy Stuart, Canadian Field Epidemiology Program, Public Health Agency of Canada; Kathryn Suh, Children’s Hospital of Eastern Ontario, Ottawa, Ont; Geoffrey Taylor, University of Alberta Hospital, Edmonton, Alta; Eva Thomas, Children’s and Women’s Health Center, Vancouver, BC; Nathalie Turgeon, Hôtel-Dieu de Québec du CHUQ, Que; Mary Vearncombe, Sunnybrook and Women’s College Health Sciences Centre, Toronto, Ont; Joseph Vayalumkal, Canadian Field Epidemiology Program, Public Health Agency of Canada; Karl Weiss, Maisonneuve-Rosemont Hospital, Montreal, Que; Alice Wong, Royal University Hospital, Saskatoon, Sask.; Dick Zoutman, Kingston General Hospital, Kingston, Ont.

22

Table 1: Source of CDAD infection among the patients enrolled in the surveillance, N = 1842. Source of CDAD

All patients N = 1842

Adults =>18 years

n = 1745 (95%)

Children 1 to < 18 years

n = 97 (5%)

Healthcare-associated 1493 (81%) 1430 (82%) 63 (65%) Nosocomial, my hospital

Nosocomial, other hospital Unknown

1421 71 1

1361 68 1

60 3 0

Community-acquired 292 (16%) 258 (15%) 34 (35%) No known healthcare contact

Admitted > 2 months ago Home care Information missing

128 97 19 48

114 85 18 41

14 12 1 7

Nursing home, household contact

57 (3%) 57 (3%) 0

23

Table 2: Incidence and rates of HA-CDAD by province or region among hospitalized patients 1 years of age and older, N= 1493. Province/Region

Cases*

Admissions

Per 1000

admissions

Patient-days

Per 10,000 patient-days

British Columbia 131 42,197 3.1 279,911 4.7 Alberta 175 79,145 2.2 386,575 4.5 Sask/Manitoba 69 25,214 2.7 184,153 3.8 Ontario 693 122,734 5.7 893,970 7.8 Quebec 283 25,610 11.1 237,794 11.9 Atlantic 142 37,549 3.8 366,243 3.9 Overall 1493 332,449 4.5 2,348,646 6.4 * Includes only the patients identified with HA-CDAD. Table 3: Mortality and fatality rates for patients with HA-CDAD at 30 days after onset of disease, N = 1493. Province/Region

Cases

All deaths

Mortality rate per 100

Directly related

Indirectly related

Fatality rate per 100

British Columbia 131 22 16.8 1 7 6.1 Alberta 175 16 9.1 1 1 1.1 Sask/Manitoba 69 10 14.5 1 0 1.5 Ontario 693 110 15.9 7 22 4.2 Quebec 283 64 22.6 20 22 14.8 Atlantic 142 15 10.6 1 1 1.4 Overall 1493 237 15.9 31 53 5.6 * Definitions: Mortality rate = death from all causes 30 days after onset of CDAD;

Fatality rate (or case fatality rate) = deaths directly or indirectly related to CDAD 30 days after onset.

24

Table 4: Description of the adult patients with HA-CDAD, N = 1430. All patients, N = 1430

18-64 years, n = 434

65 and older,n = 996

Patient characteristics

No. (%)

No. (%) No. (%)

P value

Mean age at onset ± SD (median, range)

70 ± 16 73 (18-101)

50 ± 12 (54)

79 ± 8 (78)

Male

735 (51%) 238 (55%) 497 (50%) NS

Mean length of stay before onset ± SD (median)

25 ± 50 (11)

25 ± 57 (8)

25 ± 46 (12)

NS

Type of patient Acute care

Long term care

1242 (87%) 188 (13%)

415 (96%) 19 (4%)

827 (83%) 169 (17%)

<0.0001

Admitted from Home

Another hospital Long term care facility Other

1073 (75%) 175 (12%) 127 (9%) 55 (4%)

346 (80%) 65 (15%) 11 (3%) 12 (3%)

727 (73%) 110 (11%) 116 (12%)

43 (4%)

0.0068 0.0369 <0.0001NS

Location at onset Medicine

Surgery Intensive care unit Home Long-term care Oncology/Hematology Combined Med/Surgical BMT/Transplant Unit Other

609 (43%) 327 (23%) 142 (10%)

92 (6%) 58 (4%) 53 (4%) 45 (3%) 22 (2%) 82 (5%)

130 (30%) 123 (28%) 53 (12%) 33 (8%) 5 (1%)

31 (7%) 16 (4%) 18 (4%) 25 (6%)

479 (48%) 204 (21%)

89 (9%) 59 (6%) 53 (5%) 22 (2%) 29 (3%) 4 (0.4%) 57 (6%)

<0.00010.0011 NS NS 0.0002 <0.0001NS <0.0001NS

Chronic disease * Diabetes

Heart disease Lung disease Cancer Liver disease Kidney disease Dementia Immunocompromised Other

331 (23%) 514 (36%) 327 (23%) 290 (20%)

62 (4%) 229 (16%) 101 (7%) 120 (8%)

338 (24%)

82 (19%) 76 (18%) 57 (13%)

125 (29%) 38 (9%)

63 (15%) 5 (1%)

72 (17%) 103 (24%)

249 (25%) 438 (44%) 270 (27%) 165 (17%)

24 (2%) 166 (17%) 96 (10%) 48 (5%)

235 (24%)

0.0118 <0.0001<0.0001<0.0001<0.0001NS <0.0001<0.0001NS

* May have more than one chronic disease

25

Table 5: Description of the treatments and medical interventions for adult patients with HA-CDAD, N = 1430.

All patients, N = 1430

18-64 years, n = 434

65 and older,n = 996

Patient characteristics

No. (%)

No. (%) No. (%)

P value

Initial CDAD treatment* No treatment

Metronidazole PO or IV Vancomycin Cholestyramine IVIG Probiotics

81 (6%) 1215 (85%) 230 (16%)

18 (1%) 3 (0.2%) 51 (4%)

28 (7%) 381 (88%) 52 (12%) 3 (0.7%)

0 13 (3%)

53 (5%) 834 (84%) 178 (18%)

15 (2%) 3 (0.3%) 38 (4%)

NS 0.0487 0.0053 NS NS NS

Other interventions Discontinued antibiotics

Endoscopy Surgical consult Infectious disease or GI consult

181 (13%) 51 (4%) 44 (3%) 12 (1%)

68 (16%) 21 (5%) 18 (4%) 2 (0.5%)

113 (11%) 30 (3%) 26 (3%) 10 (1%)

0.0238 NS NS NS

Initial treatment changed

246 (17%) 63 (15%) 183 (18%) NS

Reason treatment changed Failure to respond

Intolerance to antibiotic Complications Inappropriate treatment

145 (10%) 21 (2%) 12 (1%) 58 (4%)

28 (7%) 10 (2%) 2 (0.5%) 17 (4%)

117 (12%) 11 (1%) 10 (1%) 41 (4%)

0.0023 NS NS NS

* No patients received Fucidin, Bacitracin, or Linezolid; patients may be on more than one treatment

Table 6: Initial treatment for CDAD among the patients with HA-CDAD by province or region, N = 1430.

British Columbia n = 128

Alberta n = 153

Sask/Man n = 67

Ontario n = 664

Quebec n = 282

Atlantic n = 136

Initial treatment for CDAD No. (%) No. (%) No. (%)

No. (%) No. (%) No. (%)

No treatment, n = 81

6 (5%) 15 (10%) 1 (2%) 30 (5%) 19 (7%) 10 (7%)

Metronidazole, n = 1215

117 (91%) 132 (86%) 66 (99%) 622 (94%) 159 (56%) 119 (88%)

Vancomycin, n = 230

5 (4%) 15 (10%) 1 (2%) 36 (5%) 159 (56%) 14 (10%)

Other, n = 67*

2 (2%) 3 (2%) 0 37 (6%) 23 (8%) 2 (2%)

* Cholestyramine, IVIG, and/or Probiotics. Patients may be on more than one therapy.

Table 7: Frequency of adverse outcomes among the adult patients with HA-CDAD in the first 30 days following onset of disease, N = 1430.

All patients, N = 1430

18 -64 years,n = 434

65 and older, n = 996

Patient characteristics No. (%)

No. (%) No. (%)

P value

Complications from CDAD

319 (22%) 88 (20%) 231 (23%) NS

Type of complication Relapse

Bowel perforation GI bleed, transfusion Toxic megacolon Bacteraemia Dehydration Hyopkalemia Other*

125 (9%) 1 (0.1%) 16 (1%) 17 (1%) 22 (2%) 84 (6%) 39 (3%) 47 (3%)

39 (9%) 1 (0.2%) 9 (2%)

3 (0.7%) 7 (2%)

14 (3%) 11 (3%) 15 (3%)

86 (9%) 0

7 (0.7%) 14 (1%) 15 (2%) 70 (7%) 28 (3%) 32 (3%)

NS NS 0.029 NS NS 0.005 NS NS

Admitted to ICU

31 (2%) 7 (2%) 24 (2%) NS

Colectomy

12 (1%) 4 (1%) 8 (1%) NS

Death, all causes

233 (16.3%) 35 (8.1%) 198 (19.9%) < 0.0001

Death, related to CDAD† 82 (5.7%) 9 (2.1%) 73 (7.3%) < 0.0001 Directly related

Indirectly related

31 51

1 8

30 43

Severe outcome†

104 (7.3%) 17 (3.9%) 87 (8.7%) 0.0013

* Pseudomembraneous colitis and/or gastrointestinal bleed not requiring transfusions † Severe outcome defined as admission to ICU, colectomy and/or death, directly and indirectly related to CDAD. 21 patients had more than one severe outcome.

28

Table 8: Univariate analysis of the variables associated with severe outcome in patients with HA-CDAD, N = 1430.

Severe outcome n = 104

Variable

No. of

patients No. %

RR* (95% CI)

P

value

Adults 65 and older Adults 18 to 64 years of age

996 434

87 17

8.7 3.9

2.23 (1.34-3.70) Ref

0.0013

Long term care Acute care

188 1242

22 82

11.7 6.6

1.77 (1.14-2.78) Ref

0.0121

Age of the patient by decade† 90+ years

80 -89 years 70-79 years 60-69 years 50-59 years 40-49 years 18-39 years

84 311 418 275 163 86 93

11 35 27 17 5 3 6

13.1 11.3 6.5 6.2 3.1 3.5 6.5

1.92 (1.05-3.51) 1.61 (1.21-2.16) 0.88 (0.63-1.23) 0.84 (0.54-1.31) 0.40 (0.17-0.96) 0.46 (0.15-1.43) 0.88 (0.39-1.96)

0.03420.00220.44650.43830.02810.16330.7525

Admitted from Home

Another hospital Long term care facility Other

1073 175 127 55

62 18 18 6

5.7 10.3 14.2 10.9

0.49 (0.34-0.71) 1.50 (0.92-2.43) 2.15 (1.34-3.45) 1.53 (0.70-1.05)

0.00020.10130.00170.2849

Location at onset Medicine

Surgery ICU Home Long-term care Oncology/Haematology Combined Med/Surgical BMT/Transplant Unit Other

609 327 142 92 58 53 45 22 82

47 12 15 11 8 2 0 1 8

7.8 3.7

10.6 12.0 13.8 3.8 0.0 4.6 9.8

1.11 (0.77-1.61) 0.44 (0.24-0.79) 1.53 (0.91-2.57) 1.72 (0.96-3.10) 1.97 (1.01-3.86) 0.51 (0.13-2.01)

- 0.62 (0.09-4.25) 1.37 (0.69-2.72)

0.57690.00430.11160.07370.05090.42620.07161.00000.3776

* Abbreviations: RR, relative risk; CI, confidence interval; PO, per os; IV, intravenous; ID, infectious disease; GI, gastroenterology. P value: Chi-square or Fisher’s exact test where appropriate. † Comparing each category to all others

29

Table 8 (cont.): Univariate analysis of variables associated with severe outcome in patients with HA-CDAD, N = 1430.

Severe outcome n = 104

Variable

No. of

patients No. %

RR* (95% CI)

P value

Chronic disease† Diabetes

Heart disease Lung disease Cancer Liver disease Kidney disease Dementia Immunocompromised Other

331 514 327 290 62

229 101 120 338

30 41 29 18 7

17 13 8

24

9.1 8.0 8.9 6.2

11.3 7.4

12.9 6.7 7.1

1.35 (0.90-2.02) 1.16 (0.79-1.69) 1.30 (0.87-1.97) 0.83 (0.50-1.34) 1.59 (0.77-3.28) 1.09 (0.62-1.69) 1.88 (1.10-3.24) 1.00 (0.94-1.06) 0.97 (0.62-1.50)

0.1524 0.4426 0.2058 0.4337 0.2130 0.9236 0.0246 0.8500 0.8891

Initial treatment for CDAD No treatment

Metronidazole (PO or IV) Vancomycin Others

81 1215 230 67

9 75 39 5

11.1 6.2

16.9 7.5

0.97 (0.91-1.02) 0.84 (0.74-0.95) 2.60 (1.97-3.45) 1.03 (0.43-2.50)

0.1708 <0.0001<0.00010.8128

Initial treatment changed No change

246 1184

38 66

15.5 5.6

2.77 (1.90- 4.03) Ref

<0.0001

Other medical interventions Discontinued antibiotics

Surgical consult Endoscopy ID or GI consult

181 44 51 12

14 17 3

7.7 38.6 5.9

1.07 (0.64-1.78) 8.03 (4.52-14.24) 0.80 (0.25-2.51)

-

0.7978 <0.00010.6979

Province or region British Columbia

Alberta Sask/Manitoba Ontario Quebec Atlantic

128 153 67

664 282 136

10 5 4

35 47 3

7.8 3.3 6.0 5.3

16.7 2.2

1.08 (0.58-2.00) 0.43 (0.18-1.03) 0.81 (0.30-2.18) 0.71 (0.53-0.93) 2.55 (2.00-3.25) 0.29 (0.09-0.89)

0.8053 0.0436 0.6742 0.0067

<0.00010.0168

* Abbreviations: RR, relative risk; CI, confidence interval; PO, per os; IV, intravenous; ID, infectious disease; GI, gastroenterology. P value: Chi-square or Fisher’s exact test where appropriate. † Comparing each category to all others, patients may have more than one chronic disease and treatment/intervention.

30

Table 9: Molecular characterization of C.difficile isolated from the stool specimens of the patients with HA-CDAD, N = 1008.

All patients, N = 1008

18 -64 years, n = 289

65 and older, n = 719

P value Strain characteristics No. (%)

No. (%) No. (%)

PFGE Small Patterns NAP1/027*

NAP2 (J-strain) NAP3 NAP4 NAP5 NAP6 Other patterns

311 (31%) 283 (28%)

11 (1%) 36 (4%) 3 (0.3%) 31 (3%)

333 (33%)

71 (25%) 75 (26%)

3 (1%) 16 (6%) 1 (0.4%) 15 (5%)

108 (37%)

240 (33%) 208 (29%)

8 (1%) 20 (3%) 2 (0.3%) 16 (2%)

225 (31%)

0.0062 NS NS

0.0331 NS

0.0137 NS

Binary toxin+ tcdC gene deletion+

316 (31%) 77 (27%) 239 (33%) 0.0412

NAP1/027 Binary toxin+ tcdC gene deletion+

296 (29%) 68 (24%) 228 (32%) 0.0099

Table 10: Distribution of the NAP PFGE subtypes isolated in stool specimens of the patients with HA-CDAD by province or regions, N = 1008.

British Columbia

n = 81

Alberta n = 111

Sask/Man n = 37

Ontario n = 451

Quebec n = 225

Atlantic n = 103

NAP PFGE subtype

No. (%) No. (%) No. (%)

No. (%) No. (%) No. (%)

NAP1, n = 311

10 (12%) 22 (20%) 3 (7%) 100 (22%) 171 (76%) 5 (5%)

NAP2, n = 283

16 (20%) 25 (23%) 8 (22%) 157 (35%) 13 (6%) 64 (62%)

NAP3, n = 11

0 0 0 10 (2%) 1 (0.4%) 0

NAP4, n = 36

1 (1%) 3 (2%) 5 (14%) 21 (5%) 5 (2%) 1 (1%)

NAP5, n = 3

0 0 0 2 (0.4%) 0 1 (1%)

NAP6, n = 31

4 (5%) 5 (5%) 0 12 (3%) 3 (1%) 7 (7%)

All Others, n = 333

50 (62%) 56 (51%) 21 (57%) 149 (33%) 32 (14%) 25 (24%)

32

Table 11: Frequency of resistance to select antimicrobial agents of C.difficile isolated from the stool specimens of the patients with HA-CDAD by PFGE small patterns, N = 1008.

All strains, N = 1008

NAP1, n = 311

NAP2, n = 283

NAP 3, n = 11

NAP4, n = 36

NAP5, n = 3

NAP6, n = 31

All others, n = 333

Antimicrobial agent* No. (%)

No. (%) No. (%) No. (%)

No. (%) No. (%) No. (%) No. (%)

Clindamycin

868 (86%) 256 (82%) 280 (99%) 10 (91%) 22 (61%) 3 (100%) 27 (87%) 270 (81%)

Fluoroquinolone Levofloxacin

Gatifloxacin Moxifloxacin

747 (74%) 672 (67%) 673 (67%)

287 (92%) 259 (83%) 258 (83%)

276 (98%) 276 (98%) 277 (98%)

10 (91%) 10 (91%) 10 (91%)

8 (22%) 3 (8%)

4 (11%)

3 (100%) 3 (100%) 3 (100%)

8 (26%) 0 (0%) 0 (0%)

155 (47%) 121 (36%) 121 (36%)

Cephalosporin Cefazolin

Ceftriaxone

957 (95%) 658 (34%)

310 (99%) 245 (79%)

283 (100%) 276 (98%)

9 (82%) 7 (63%)

35 (97%) 10 (28%)

3 (100%) 3 (100%)

29 (94%) 3 (10%)

288 (86%) 114 (34%)

* No strains were resistant to Metronidazole, Vancomycin or Teicoplanin. All strains were resistant to Ciprofloxacin, Cefuroxime and Cefotaxime.

Table 12: Frequency of severe outcomes among the adult patients with HA-CDAD in the first 30 days following onset of disease according to the strain type, N = 1008.

Death,* n = 66

Severe outcome,

n = 80

Strain characteristics

All,

N = 1008No. (%)

RR (95%CI)

P value No. (%)

RR (95%CI)

P value

NAP1/027 All others

311 697

34 (10.9) 32 (4.6)

2.34 (1.50-3.79) Ref

<0.0001 39 (12.5%) 41 (5.9%)

2.13 (1.40-3.24) 0.0003

NAP2 (J-strain) Others including NAP1

283 725

9 (3.2) 57 (7.7)

0.41 (0.20-0.81) Ref

0.0069 13 (4.6%) 67 (9.2%)

0.49 (0.28-0.86) 0.0142

NAP1/Binary toxin/tcdC deletion Others

296 712

34 (11.4) 32 (4.5)

2.55 (1.50-4.11) Ref

<0.0001 38 (12.8) 42 (5.9)

2.17 (1.33-3.35) Ref

0.0002

* Death directly or indirectly related to HA-CDAD

12 207

22

76

5

88 8093

78

24

95

0%

25%

50%

75%

100%

BC Alberta Sask/Man Ontario Quebec Atlantic

Perc

enta

ge N

AP1

/027

vs.

Oth

ers

NAP1/027 Others

Figure 1: Distribution of C. difficile NAP1/027 in adults with HA-CDAD, by province or region (n=1008)

35

0

4

8

12

16

20

24

18-39 40-49 50-59 60-69 70-79 80-89 90+

Age (years)

Perc

ent s

ever

e C

DA

D

NAP1/027 Non-NAP1

NSNS

NS

P=0.03P=0.13

P=0.001

NS

Figure 2: Effect of strain type on severe outcomes, by age, N =1008